Skip to content
conftrace_

monosemantic feature

3 papers

Explore in graph

Co-occurring keywords · papers mentioning both

sparse autoencoder (3) polysemous word (1) llm interpretability (1) low-rank adaptation (1) parameter-efficient fine-tuning (1) reward model (1) safety alignment (1) feature attribution (1) semantic representation (1)

Papers

Interpretable Reward Model via Sparse Autoencoder AAAI 2026 Interpretable Safety Alignment via SAE-Constructed Low-Rank Subspace Adaptation ACL 2026 Rethinking Evaluation of Sparse Autoencoders through the Representation of Polysemous Words ICLR 2025